Evil Lyn is one of the most formidable villains in the Masters of the Universe universe, serving as Skeletor’s chief strategist and magical powerhouse. Intelligent, ambitious, and fiercely independent, she is far more than a sidekick, often pursuing her own goals when it suits her. Her sharp wit and calculating nature make her just as dangerous as her sorcery, and she carries herself with a cold confidence that sets her apart from the more reckless villains around her. Whether commanding dark magic or maneuvering behind the scenes, Evil Lyn remains one of Eternia’s most compelling and enduring antagonists. Anima-Base <span style="background-color:rgb(26, 27, 30);color:rgb(193, 194, 197);font-family:-apple-system, BlinkMacSystemFont, &quot;Segoe UI&quot;, Roboto, Helvetica, Arial, sans-serif, &quot;Apple Color Emoji&quot;, &quot;Segoe UI Emoji&quot;;font-size:16px">Anima Base version of Evil Lyn. Tags are the same but this version was trained on 100% images from the original cartoon and nothing else. So if you don't tag an artist style or use a lora in addition to this one, you will get the old 80's Filmation style cartoon. Z-Image Turbo:I probably spent more time on this LoRa than my other models. I wanted the model to have flexibility across different styles. I ran into a few issues. First was a consistent outfit. Depending on the source (original, movie, netflix series), Evil Lyn's look varies quite a bit. I chose to focus only on her original outfit. But that also severely reduced the amount of resources as far as images. So I procured what I could find, upscaled the cartoon stills using SeedVR2, then took a handful of images and ran them through Klein 9b to get different media styles to give the data set more variety. The majority of the dataset consists still of screenshots from the original cartoon. So the model will defer to that style most of the time. Not always but most. The captioning was just focused on her outfit and characteristics. Which also makes those tags important to mention if you want them. If you don't want something, leave it out or chose to replace it and I believe the Lora has the flexibility to do that. Changing the fabric to metallic on the body suit and helmet is easy and responsive. I trained the model on Runpod with AI-toolkit. Tips: Like I said previously, in order to achieve her original outfit, the Lora responds well to high detail. So for her original outfit, try: "black cap with purple and blue headpiece" "purple paneled leotard suit" "black cape" "blue forearm cuffs" "purple mid calf boots". All of these tags can be modified as well. like "glossy metallic and chrome black cap with purple and silver headpiece"

Evil Lyn

Anima Base version of Evil Lyn. Tags are the same but this version was trained on 100% images from the original cartoon and nothing else. So if you don't tag an artist style or use a lora in addition to this one, you will get the old 80's Filmation style cartoon.

Files with SHA256: 2668e26fc503de5c88c7a7f3de081e9c8b25231d48f8a412b8c2fbb886f3e300